Arnaud Giersch [Mon, 18 Feb 2013 10:16:53 +0000 (11:16 +0100)]
Also set MPI_ERROR in smpi_empty_status.
Arnaud Giersch [Fri, 15 Feb 2013 16:17:44 +0000 (17:17 +0100)]
Remove unused script.
Takahiro Hirofuchi [Fri, 15 Feb 2013 12:48:01 +0000 (13:48 +0100)]
add the definition of VM state to the surf layer
FIXME: Where should the VM state be defined? Only in the surf layer, or
also in the msg layer?
Takahiro Hirofuchi [Fri, 15 Feb 2013 12:34:21 +0000 (13:34 +0100)]
add the cpu model object in the workstation model object
Takahiro Hirofuchi [Fri, 15 Feb 2013 12:17:45 +0000 (13:17 +0100)]
fix minor issues aroud the msg layer
Takahiro Hirofuchi [Fri, 15 Feb 2013 11:51:53 +0000 (12:51 +0100)]
refine code around model types and objects
The model_type member of an action object is renamed model_obj. This
change makes it clear that this member points to a surf model objet, not
the type of a model.
The model type member is added to a surf model object. This allows us to
know what type a model object is and what extension field is accessible.
Takahiro Hirofuchi [Thu, 14 Feb 2013 17:28:14 +0000 (18:28 +0100)]
add two cpu model objects for physical and virtual
We delete surf_cpu_model, which was only one cpu model object on the
system. Now, we add surf_cpu_model_pm and surf_cpu_model_vm for the
physical and virtual machine layers, respectively.
FIXME: cpu_ti will not work correctly because using global variables. We
have to update cpu_ti to support multiple cpu model objects.
Augustin Degomme [Fri, 15 Feb 2013 09:39:44 +0000 (10:39 +0100)]
be case insensitive when trying to change main name in F90 files
Augustin Degomme [Fri, 15 Feb 2013 09:31:14 +0000 (10:31 +0100)]
add mpi_testall binding for fortran
Arnaud Giersch [Fri, 15 Feb 2013 08:55:37 +0000 (09:55 +0100)]
Cosmetics: add space before dot.
Arnaud Giersch [Thu, 14 Feb 2013 22:49:39 +0000 (23:49 +0100)]
Using memcpy looks a bit overkill here.
Arnaud Giersch [Thu, 14 Feb 2013 22:45:29 +0000 (23:45 +0100)]
Remove spurious brace from error message.
Takahiro Hirofuchi [Thu, 14 Feb 2013 16:49:55 +0000 (17:49 +0100)]
add a model object to the arguments of callbacks
For VM support, we have multiple cpu model objects on the system. The
object of a surf model may not be only one on the system. Thus, we have
to modify the code that assumes the system has only one global object of
each surf model.
We add a model object to the first argument of the callbacks of struct
surf_model_private; i.e., share_resources(), update_actions_state(), and
finalize(). In each callbacks, we delete the use of the global model
object.
Arnaud Giersch [Thu, 14 Feb 2013 16:28:13 +0000 (17:28 +0100)]
Remove remaining traces from synchronized dynars.
Had to add some #include "xbt/synchro_core.h".
Arnaud Giersch [Thu, 14 Feb 2013 16:05:41 +0000 (17:05 +0100)]
Increase tesh timeouts for slow machines.
Arnaud Giersch [Thu, 14 Feb 2013 16:05:08 +0000 (17:05 +0100)]
Make test reproducible on slow machines.
Takahiro Hirofuchi [Thu, 14 Feb 2013 15:54:35 +0000 (16:54 +0100)]
add model_list_invoke used for solving a problem
The callback function (i.e., share_resouces) of cpu_model and
network_model is now called from those of these workstation models.
They are not directly called from surf_solve(). model_list_invoke will
contain only surf_workstation and surf_vm_workstation.
- xbt_dynar_t model_list: for destroying all models correctly
- xbt_dynar_t model_list_invoke: for invoking the callback
FIXME: There might be other callbacks that should be called in the
iteration of model_list_invoke, not model_list. Update code if
neccessary.
Arnaud Giersch [Thu, 14 Feb 2013 15:07:26 +0000 (16:07 +0100)]
Csmetics: algin braces.
Arnaud Giersch [Thu, 14 Feb 2013 14:45:24 +0000 (15:45 +0100)]
Remove declarations for undefined functions.
Arnaud Giersch [Thu, 14 Feb 2013 14:28:03 +0000 (15:28 +0100)]
Use xbt_XXX_is_empty() instead of testing xbt_XXX_length() against 0.
XXX is 'dict' or 'dynar'.
Arnaud Giersch [Thu, 14 Feb 2013 14:26:02 +0000 (15:26 +0100)]
Simply say func(...) instead of (*func)(...) when func is a pointer to function.
Arnaud Giersch [Thu, 14 Feb 2013 14:25:47 +0000 (15:25 +0100)]
Use CATCH_ANONYMOUS whenever possible.
Arnaud Giersch [Thu, 14 Feb 2013 14:25:43 +0000 (15:25 +0100)]
Replace, when possible, calls to bprintf() by calls to xbt_strdup().
Arnaud Giersch [Thu, 14 Feb 2013 14:14:50 +0000 (15:14 +0100)]
Also comment declaration for now unused variable.
Arnaud Giersch [Thu, 14 Feb 2013 14:14:09 +0000 (15:14 +0100)]
Remove unused static functions.
Arnaud Giersch [Thu, 14 Feb 2013 14:10:34 +0000 (15:10 +0100)]
Remove declarations for undefined functions.
Arnaud Giersch [Thu, 14 Feb 2013 14:19:28 +0000 (15:19 +0100)]
Declaration for static functions in header file does not make much sense.
Lucas Schnorr [Thu, 14 Feb 2013 13:41:11 +0000 (14:41 +0100)]
[trace] fix tesh, changes only in event ordering, same number of links kept
Jonathan Rouzaud-Cornabas [Thu, 14 Feb 2013 13:26:18 +0000 (14:26 +0100)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Jonathan Rouzaud-Cornabas [Thu, 14 Feb 2013 13:26:15 +0000 (14:26 +0100)]
Fix recv/irecv with bounded rate
Lucas Schnorr [Thu, 14 Feb 2013 13:20:42 +0000 (14:20 +0100)]
[trace] fix simdag tracing tesh (change in event ordering only)
Lucas Schnorr [Thu, 14 Feb 2013 13:13:56 +0000 (14:13 +0100)]
[trace] remove useless if, fix identation
Lucas Schnorr [Thu, 14 Feb 2013 13:12:32 +0000 (14:12 +0100)]
[trace] correct place to keep route allocation + free
Lucas Schnorr [Thu, 14 Feb 2013 12:59:41 +0000 (13:59 +0100)]
[trace] add debug messages to debug graph extraction in network models
- only in generic_get_graph for now
Lucas Schnorr [Thu, 14 Feb 2013 12:58:54 +0000 (13:58 +0100)]
[trace] cosmetics and typos
Lucas Schnorr [Thu, 14 Feb 2013 12:58:35 +0000 (13:58 +0100)]
[trace] useless declaration
Jonathan Rouzaud-Cornabas [Thu, 14 Feb 2013 12:59:13 +0000 (13:59 +0100)]
New version of recv/irecv with bounded rate
Takahiro Hirofuchi [Thu, 14 Feb 2013 12:53:35 +0000 (13:53 +0100)]
Merge branch 'hypervisor' of scm.gforge.inria.fr:/gitroot/simgrid/simgrid into hypervisor
The missing header file was added.
alebre [Thu, 14 Feb 2013 12:53:00 +0000 (13:53 +0100)]
Add workstation_private.h
Jonathan Rouzaud-Cornabas [Thu, 14 Feb 2013 12:03:05 +0000 (13:03 +0100)]
Remove commented code from surf_routing_cluster
Jonathan Rouzaud-Cornabas [Thu, 14 Feb 2013 12:01:56 +0000 (13:01 +0100)]
Remove commented code from surf_routing_cluster
Arnaud Giersch [Wed, 13 Feb 2013 17:14:22 +0000 (18:14 +0100)]
Check if INCLUDE_DIRECTORIES were found.
Arnaud Giersch [Wed, 13 Feb 2013 16:53:43 +0000 (17:53 +0100)]
Add some debug messages.
Arnaud Giersch [Wed, 13 Feb 2013 16:37:25 +0000 (17:37 +0100)]
Add option -X to less, to avoid to clear the screen.
Arnaud Giersch [Wed, 13 Feb 2013 16:27:46 +0000 (17:27 +0100)]
Add missing files to the archive.
Arnaud Giersch [Wed, 13 Feb 2013 09:54:46 +0000 (10:54 +0100)]
Fix errors caught by java -Xcheck:jni.
Fixed warnings:
WARNING in native method: JNI call made with exception pending
Arnaud Giersch [Wed, 13 Feb 2013 09:49:32 +0000 (10:49 +0100)]
More long -> intptr_t changes.
Arnaud Giersch [Wed, 13 Feb 2013 09:19:10 +0000 (10:19 +0100)]
Add missing class initializer for VM.java.
Augustin Degomme [Wed, 13 Feb 2013 09:51:39 +0000 (10:51 +0100)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
suter [Wed, 13 Feb 2013 09:51:08 +0000 (10:51 +0100)]
word was missing, rephrase.
Martin Quinson [Wed, 13 Feb 2013 08:07:59 +0000 (09:07 +0100)]
add another word about the trap of dynar_search with pointed elements
Arnaud Giersch [Tue, 12 Feb 2013 23:52:41 +0000 (00:52 +0100)]
Typos: binded -> bound.
Martin Quinson [Wed, 13 Feb 2013 00:02:54 +0000 (01:02 +0100)]
document that it's safe to interrupt a foreach with a break or return statement
Martin Quinson [Wed, 13 Feb 2013 00:00:34 +0000 (01:00 +0100)]
document that the search function on dynar don't work well with pointed values
Arnaud Giersch [Tue, 12 Feb 2013 22:50:01 +0000 (23:50 +0100)]
Set correct NATIVE folder for 32bits x86.
Arnaud Giersch [Tue, 12 Feb 2013 22:26:55 +0000 (23:26 +0100)]
Set cmake_minimum_version for the Java bindings.
UseJava.cmake don't exist before cmake 2.8.6.
Arnaud Giersch [Tue, 12 Feb 2013 21:22:04 +0000 (22:22 +0100)]
Move Java build rules in a separate cmake file.
Arnaud Giersch [Tue, 12 Feb 2013 20:53:53 +0000 (21:53 +0100)]
Fix build warnings on 32bits archs.
Use intermediate type intptr_t for conversions between pointer
types and jlong.
Arnaud Giersch [Tue, 12 Feb 2013 16:48:16 +0000 (17:48 +0100)]
Try to make the smpi-mpich-env test more robust in multitasking environments.
Only select the processes that share the same session ID.
Arnaud Giersch [Tue, 12 Feb 2013 15:08:08 +0000 (16:08 +0100)]
Initialize variable, and make gcc happy.
Arnaud Giersch [Tue, 12 Feb 2013 15:06:13 +0000 (16:06 +0100)]
Reindent.
Arnaud Giersch [Tue, 12 Feb 2013 14:58:57 +0000 (15:58 +0100)]
Kill unused variable.
Arnaud Giersch [Tue, 12 Feb 2013 14:57:52 +0000 (15:57 +0100)]
Add missing declaration.
Plus whitespace cleanup.
Augustin Degomme [Tue, 12 Feb 2013 14:27:31 +0000 (15:27 +0100)]
issend should be treated as isend, even if not detached
Augustin Degomme [Tue, 12 Feb 2013 14:15:33 +0000 (15:15 +0100)]
correct behavior of MPI_Send which used to values of MPI_Isend timing factors
Arnaud Giersch [Tue, 12 Feb 2013 13:31:14 +0000 (14:31 +0100)]
Update .gitgnore.
Match any build directory named "build_*", and add etags' tag table.
Arnaud Giersch [Tue, 12 Feb 2013 13:20:04 +0000 (14:20 +0100)]
Tests msg-bittorrent-*-parallel are not broken anymore.
Arnaud Giersch [Tue, 12 Feb 2013 13:14:50 +0000 (14:14 +0100)]
Fix the parallel execution for the bittorrent example.
The problem was that RngStream creation is not thread safe. Furthermore,
each created RngStream is dependent of the previously created RngStreams.
The problem is circumvented by creating the RngStreams before the launch
of the application. Each host has its own stream, that can be retrieved
through MSG_host_get_data().
Arnaud Giersch [Tue, 12 Feb 2013 00:10:49 +0000 (01:10 +0100)]
Fix double free error.
Error reported by Borja Bergua, and patch inspired by Martin
Quinson's proposal.
Jonathan Rouzaud-Cornabas [Tue, 12 Feb 2013 13:39:29 +0000 (14:39 +0100)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
Jonathan Rouzaud-Cornabas [Tue, 12 Feb 2013 13:39:05 +0000 (14:39 +0100)]
Bounded receive
Takahiro Hirofuchi [Tue, 12 Feb 2013 13:13:40 +0000 (14:13 +0100)]
remove the use of surf_cpu_model->func
Takahiro Hirofuchi [Tue, 12 Feb 2013 12:38:44 +0000 (13:38 +0100)]
add minor cleanup around properties
We will revisit here after the surf layer is updated.
alebre [Tue, 12 Feb 2013 10:46:51 +0000 (11:46 +0100)]
Additional comments - Adrien/Takahiro
Augustin Degomme [Mon, 11 Feb 2013 17:27:33 +0000 (18:27 +0100)]
add an smpi/ois factor, to account for the fact that MPI_Send and MPI_Isend internal times may be differents
Augustin Degomme [Mon, 11 Feb 2013 17:22:17 +0000 (18:22 +0100)]
activate a bunch of tests using mpi_ssend, and change back those where they had been changed to mpi_send
Augustin Degomme [Mon, 11 Feb 2013 16:43:44 +0000 (17:43 +0100)]
add support for Ssend/Issend in SMPI
Augustin Degomme [Mon, 11 Feb 2013 15:33:26 +0000 (16:33 +0100)]
Revert "always have display_timing activated for smpi, as it is quite useful"
This reverts commit
f641b68a3c807956396f8f70965ce9c4644dcfa5.
Augustin Degomme [Mon, 11 Feb 2013 14:50:16 +0000 (15:50 +0100)]
add option to set the threshold where smpi send stops behaving like isend.
This was hardcoded to 65536, but experiments show that sometimes 327680 might be better
Augustin Degomme [Mon, 11 Feb 2013 14:49:08 +0000 (15:49 +0100)]
small change for smpi win
Augustin Degomme [Mon, 11 Feb 2013 14:46:59 +0000 (15:46 +0100)]
bug : smpi tracing options were replacing those the user had set
Augustin Degomme [Tue, 5 Feb 2013 17:27:08 +0000 (18:27 +0100)]
small fixes to repair unix build
Augustin Degomme [Tue, 5 Feb 2013 17:19:47 +0000 (18:19 +0100)]
rename main function for windows
Augustin Degomme [Tue, 5 Feb 2013 17:19:27 +0000 (18:19 +0100)]
several changes in the way cmake handles windows
Augustin Degomme [Mon, 11 Feb 2013 15:19:13 +0000 (16:19 +0100)]
resolve cmakelists conflict
Augustin Degomme [Tue, 5 Feb 2013 17:16:09 +0000 (18:16 +0100)]
remove a few unused things for win, but we have more bugs after ..
Augustin Degomme [Tue, 5 Feb 2013 17:14:38 +0000 (18:14 +0100)]
various fixes to hopefully compile under win32
Augustin Degomme [Mon, 11 Feb 2013 14:50:39 +0000 (15:50 +0100)]
Merge branch 'master' of git+ssh://scm.gforge.inria.fr//gitroot/simgrid/simgrid
suter [Fri, 8 Feb 2013 15:13:17 +0000 (16:13 +0100)]
handle new action trace files properly
suter [Fri, 8 Feb 2013 15:12:35 +0000 (16:12 +0100)]
add three more tests to smpi_replay : alltoall, alltoallv and waitall
suter [Fri, 8 Feb 2013 15:11:53 +0000 (16:11 +0100)]
new traces files were added in examples/smpi/replay. treat them as the
others
George Markomanolis [Fri, 8 Feb 2013 12:02:24 +0000 (14:02 +0200)]
Fixing a possible bug, I should use parse_double() instead of atoi() if I want to handle values like 5e8. Adding examples for MPI_Alltoallv, MPI_Alltoall, MPI_Bcast/MPI_Reduce with various datatypes and MPI_Waitall.
suter [Fri, 8 Feb 2013 09:56:07 +0000 (10:56 +0100)]
rewriting
suter [Fri, 8 Feb 2013 09:52:28 +0000 (10:52 +0100)]
revalidate tesh file as the reduce action has been modified
Arnaud Giersch [Thu, 7 Feb 2013 22:13:55 +0000 (23:13 +0100)]
Make this part automatically generated.
And avoid to give a direct link to the archive, since it's too error prone.
Arnaud Giersch [Thu, 7 Feb 2013 21:54:51 +0000 (22:54 +0100)]
Typo.
Martin Quinson [Thu, 7 Feb 2013 20:15:21 +0000 (21:15 +0100)]
Fix the download link in the install documentation
Martin Quinson [Thu, 7 Feb 2013 20:10:50 +0000 (21:10 +0100)]
ops, didn't knew how gforge downloads work, actually
Martin Quinson [Thu, 7 Feb 2013 01:18:30 +0000 (02:18 +0100)]
a more welcoming README file
Martin Quinson [Thu, 7 Feb 2013 01:14:02 +0000 (02:14 +0100)]
Seed the random generators before the main()
Otherwise, it is possible that the user set the seed s/he wants before
initializing simgrid, that would then override it with another seed.
That would be a *very mean* side effect of initializing simgrid :)